Integrative systematics at the species level: plumage, songs and molecular phylogeny of quailfinches Ortygospizan

Robert B. Payne & Michael D. Sorenson 4 Bull. B.O.C. 2007 127(1)

Integrative systematics at the species level:
plumage, songs and molecular phylogeny
of quailfinches Ortygospiza

by Robert B. Payne & Michael D. Sorenson

Received 20 February 2006

Species delimitations in birds generally are readily discerned and find a consensus
among ornithologists, and this happy circumstance follows from the mutually
consistent evidence that is typically derived from application of different species
concepts to the same set of birds. First, under a ‘biological species concept’,
populations that share unique behaviours and interbreed where their ranges meet are
considered the same species. Breeding in sympatry without interbreeding provides
evidence of reproductive isolation of two populations and thus of species, whereas
the occurrence of birds of intermediate morphology suggests recent or current gene
flow and indicates a single species is involved (Mayr 1963, 2000, de Queiroz 2005).
Second, a ‘phylogenetic species concept’ uses the presence of exclusive sets of
characters of birds (Cracraft 1983, Sites & Marshall 2004). Operationally, under this
concept species have been recognised based on morphological characters, much as
in Linnaeus (1758), without any phylogenetic analysis having been performed
(Wheeler & Platnick 2000). Finally, molecular data have been analysed in a
phylogenetic perspective, with the ideal of genetically exclusive lineages as a
criterion for recognising species; that is, the historical isolation and independent
evolution of populations, in the current version of a lineage or ‘evolutionary species
concept’. We now know, however, that gene trees do not always coincide with
species trees, owing to incomplete lineage sorting of genes (de Queiroz 1998, 2005,
Arbogast et al. 2002, Avise 2004), and species can be delimited without reciprocally
monophyletic gene trees (Knowles & Carstens in press).

An ‘integrative species concept’ combines these views of species in the past and
present. A complementation of independent lines of evidence is very important for
diagnosing biological species. We consider both the geographical patterns of
morphological differences between populations and the phylogeny of genetic
lineages. In addition to morphological variation and genetic lineages, we note that
songs are important to the birds: experimental evidence points to songs as a major
behavioural cue in mate recognition by breeding females (Searcy & Yasukawa 1996,
Payne et al. 2000, Sorenson et al. 2003). In consequence we interpret song
differences between morphologically recognised taxa as evidence that more than
one species may be present, and the lack of song differences as an argument that
populations are conspecific (e.g. Alström & Ranft 2003). In addition, an integrative
species concept offers a response to claims that species are best recognised simply
in terms of genetic distances between populations, insofar as rates of genetic
divergence differ between lineages, and as gene trees may coalesce only after a

Robert B. Payne & Michael D. Sorenson 5 Bull. B.O.C. 2007 127(1)

speciation event (Moritz & Cicero 2004, Dayrat 2005, Will et al. 2005). Integrative
systematics proposes that taxa should be compared within an estimated phylogeny,
rather than simply in character lists of taxa. The integration recognises that genetic
sequence data viewed in a phylogeny is part of a comprehensive view in which
geographic variation, morphology and behaviour provide equally useful information
at the species level.

African quailfinch Ortygospiza are small terrestrial finches that occur in short-
grass and seasonally flooded riverine plains of open country in sub-Saharan Africa.
They are inconspicuous, staying on the ground, then rising on whirring wings and
with rattling flight-calls. Adults are sexually dimorphic in plumage, and they vary
in bill and plumage colour and the intensity of plumage markings, especially in
males. As evidenced by plumage and by molecular data, Ortygospiza are most
closely related to the African estrildid genera Amadina, Amandava and perhaps
Paludipasser (Sorenson & Payne in Fry 2004, Sorenson et al. 2004). The number
of quailfinch species has been less certain. Here we describe the adult plumage and
bill colour of quailfinch taxa, and the evidence of breeding sympatry. We compare
songs and nestling mouth colours and patterns to evaluate any behavioural
differences between populations, both as traits that may be important in successful
reproduction and as markers of gene flow. Finally, we use molecular genetics to
determine the phylogenetic relationships among geographic populations across
Africa. We consider these criteria together to assess whether previously described
taxa represent species. Using these criteria we reason that quailfinch are best
recognised as a single species, Ortygospiza atricollis.

Systematic history and plumage variation in Ortygospiza

Based on geographic variation in plumage, previous authors have recognised one,
two or three quailfinch species. Sclater (1930a) and Chapin (1954) recognised a
single species, and Wolters (1975, 1985) recognised one species with three groups:
a west African ‘black-faced quailfinch’ O. atricollis (Vieillot, 1817), a central
African ‘black-chinned quailfinch’ O. gabonensis Lynes, 1914, and an eastern and
southern ‘African quailfinch’ O. fuscocrissa Heuglin, 1863. More recently, Fry
(2004) recognised these ‘racial groups’ as three species and reported areas of
geographic overlap between them.

Most earlier and some recent accounts recognised two quailfinch species, with
the taxa combined in different ways. Over much of Africa, white-chinned
quailfinch, described from Senegal (O. atricollis), have some white around the eye,
whereas black-chinned birds (O. gabonensis) lack this. Immelmann et al. (1965,
1977a), Mayr et al. (1968), Benson et al. (1971), Goodwin (1982) and Dickinson
(2003) recognised these as two species. In another representation of two species,
Sharpe (1890) and Shelley (1905) recognised one species with white on the throat
and around the eye, ‘O. polyzona’ (Temminck, 1823), and a second species without
white except ‘a few whitish plumelets round the eye’, O. atricollis. In a third model,
White (1963) recognised O. fuscocrissa for the distinctly spectacled forms of

Robert B. Payne & Michael D. Sorenson 6 Bull. B.O.C. 2007 127(1)

eastern and southern Africa, and O. atricollis for the west African birds and the
black-chinned birds from central Africa.

Differences in delimiting quailfinch species stem from inaccuracies in plumage
descriptions, puzzling original descriptions, and questionable records of breeding
sympatry of populations. First, the quailfinch with white around the eye have an
incomplete eye-ring, the posterior part of the ring is variably complete and the
anterior ends of the broken ring extend from the eye to the bill as white lines above
and below the lores (which vary from grey to black). These white markings are most
prominent in birds from east and southern Africa, which have a distinctive
spectacled appearance, more so than birds in west Africa. Descriptions of black- and
white-chinned quailfinch in Sharpe (1890) overlooked the presence of a small white
chin patch in the western birds. Vieillot (1817) described the small white chin patch
in his species Fringilla atricollis from Senegal, as mentioned also by Cassin (1860).
Confusion more importantly traces to Temminck’s description of Fringilla polyzona
with two specimens, a female and a male, apparently from two geographic sources.
(Temminck, 1823: col. 221, fig. 3) illustrated a pale female, and his text described
a female with the chin white and the underparts pale with the dark bars on the flanks
broadly separated by the belly. Temminck’s text description of a male, however, was
of a dark bird with a black throat. Sclater (1930a) recognised one species of
quailfinch, O. atricollis, with seven subspecies, one being O. a. polyzona
(Temminck, 1823) which Sclater (1930a: 784) recognised as being like the pale
birds in South Africa; his footnote remarked that Temminck’s description of the
male involved the Gambian form. Temminck reported the birds as from ‘les
provinces de royaume de Gambie sur les côte d’occidentales d’Afrique’. His
illustration depicts a female with a white streak above the eye; syntype RMNH
90327 has the face nearly all white and unfeathered, apparently due to feather loss
post-collection and to application of a white substance, perhaps a preservative (for
museum acronyms see Acknowledgements). No other syntypes of polyzona have
been traced; either in RMNH or in MNHN (RBP; J.-F. Voisin in litt. 2006). Roberts
(1930) declared polyzona to be a synonym of O. a. atricollis, on the grounds that the
description of the male must have priority. Grant & Mackworth-Praed (1956)
concurred and emphasised that Vieillot designated the birds as being from The
Gambia, making O. polyzona (Temminck, 1823) a synonym of O. atricollis
(Vieillot, 1817), and recent accounts have followed this reasoning (e.g., Mayr et al.
1968). Temminck’s illustration resembles both the female syntype of polyzona
(incorrectly labeled as from Senegal), and a female quailfinch (UMMZ 211483)
from the Save River near Beira, southern Mozambique.

In the dark-plumaged quailfinches, O. gabonensis Lynes, 1914, was described
from Gabon and diagnosed by the back-feathers being streaked (not uniform), the
absence of white on the chin and around the eye, and the white bars of the
underparts being broader than in other quailfinch (the ‘female’ in his description
was a juvenile: Cowles 1957). Lynes subsequently collected another new quailfinch
at Kawambwa, north-east Zambia, O. a. fuscata Sclater, 1930, the plumage nearly

Robert B. Payne & Michael D. Sorenson 7 Bull. B.O.C. 2007 127(1)

black above, dark cinnamon on the belly and narrower white bars on the underparts
than the most similar form, the dark-plumaged, black-faced O. a. ansorgei. O. a.
fuscata had the bill orange with sepia on the tip and around the nostrils (Sclater
1930b). Sclater (1930a) recognised gabonensis as a subspecies of O. atricollis, as
did Bannerman (1949). The other black-chinned taxon was O. a. dorsostriata,
described by van Someren (1921a) from western Uganda as being like gabonensis,
but ‘richer rufous on the breast; moreover, the male has a small white chin-spot, the
female not. There is no white ring round the eye.’ These dark-plumaged quailfinch
occur in central Africa, mainly at the fringes of the rainforest zone.

In the quailfinches with little white on the face and chin, other taxa have been
described in addition to nominate O. a. atricollis. O. a. ansorgei Ogilvie-Grant,
1910, from Guinea-Bissau, was described on the basis of the black chin and throat
extending onto the chestnut breast, the white bars below fewer, the upperparts
darker; one of two males had a small white patch on the chin, and in both specimens
of the type series the bill was dark red-brown above and crimson-lake below. O. a.
ansorgei in The Gambia, Guinea-Bissau, Guinea, Sierra Leone to Liberia and Côte
dIvoire in the far west, have a small white line on the chin (sometimes lacking;
Gatter 1997: 280 in Liberia; MCZ 153629 from Guinea-Bissau), and a black face
with little or no white around the eye and lores. Elsewhere in west Africa, from
Senegal and Mali to Nigeria and Cameroon, O. a. atricollis has white feathers above
and below the eye and lores. Although colour plates in regional field guides and
other works illustrate west African birds as lacking white around the eye and lores
(Serle & Morel 1977, Clement 1993, Barlow & Wacher 1997, Borrow & Demey
2001, Fry 2004), and Bates (1930) mentioned no white around the eye and lores,
nominate O. atricollis does have some white in these areas, but this is not obvious
in poorly prepared specimens. In Nigeria, eight of ten males photographed by RBP
at Bukuru, Nigeria, in September—October 1995, had a few white feathers below the
eye and on the lores (Fig. 1); five of six adult females also had some white in these
areas (mainly on the lower branch of the lores) and a partial ring below the eye.
These ‘white-chinned’ ansorgei and atricollis are otherwise dark, more like fuscata
in northern Zambia than the paler quailfinch of east and southern Africa.

O. a. ugandae van Someren, 1921, in Uganda and the North Kavirondo region
of western Kenya, was described as similar to O. a. ansorgei but uniformly grey-
brown above. Later, van Someren (1922) noted ugandae to have ‘uniform
grey-brown mantles, black foreheads, extensive black throats, and small white chin-
spots, with a white ring round the eyes; breasts pale brownish’; and dorsostriata to
be like gabonensis but ‘richer rufous below and the flanks darker. The female has
no white chin-spot. The male has a small indication of white on the chin, but no
white round the eye.’ Birds in Uganda and Sudan are intermediate between O. a.
atricollis and O. a. muelleri (ugandae are darker chestnut below than muelleri);
Sudan birds (SMNS series) have more white on the face than O. a. atricollis, in
contrast to the evaluation by Nikolaus (1987). In fact, the plumage of ugandae is
barely separable from the plumage of muelleri except for the narrower white eye-

Robert B. Payne & Michael D. Sorenson 8 Bull. B.O.C. 2007 127(1)

Figure 1. Plumage variation in west African quailfinch. All are males, except f = female. (a—f) Jos,
Nigeria, October-November 1995 (a, -/y; b, UMMZ 233845; c, -/o; d, UMMZ 233846; e, -/r; f, -/G); (g)
captive UMMZ 232576 (the specimen in Groth 1998); (h) Marakissa, The Gambia, September 1996; (1)
Ngaoundere, Cameroon, male taken with four fledglings, UMMZ 232472.

ring in ugandae, though the white eye-ring is distinct in the holotype, FMNH
257709, taken near O. a. muelleri in southern Kenya. The dry woodland and steppe
region of sub-Saharan Africa between Senegal and Sudan and into northern Uganda
and western Kenya is a nearly continuous vegetation zone (Keay 1959, Moreau
1966). This region is separated by drier country from other vegetation zones where
quailfinch occur, and we refer to the region where atricollis, ansorgei and ugandae
occur as west Africa.

O. a. fuscocrissa Heuglin, 1863, in Ethiopia, north-east Africa, has broad white
spectacles, the white lines conspicuously broader than in west African quailfinch. In
O. a. fuscocrissa the median breast and flanks have black bars broader than the
white bars, and the back is brown, more distinctly streaked blackish than in O. a.
atricollis.

Next, O. a. muelleri Zedlitz, 1911, in east Africa is similar to fuscocrissa but the
upperparts are nearly uniform with darker, indistinct streaks. The widespread O. a.

Robert B. Payne & Michael D. Sorenson 9 Bull. B.O.C. 2007 127(1)

muelleri occurs from east to southern Africa. In specimens we find little difference
between plumage in east Africa (Tanzania) and southern Africa (southern Zambia,
Zimbabwe and South Africa). O. a. bradfieldi Roberts, 1929, in Namibia, ‘grayer
and less brown’ than South African quailfinch, does not consistently differ between
these regions, and as in White (1963) and Immelmann et al. (1965, 1977a),
bradfieldi is considered a synonym of O. a. muelleri. In South Africa, O. a. digressa
Clancey, 1958, specimens are mostly darker than O. a. muelleri from south-central
Zambia and east Africa, as in Clancey (1977). Nevertheless, not all South African
specimens are darker than O. a. muelleri from south-central Zambia and east Africa;
the dark specimens from Transvaal are worn and soiled (MCZ) when compared with
birds in fresh plumage from the same areas (USNM), and these fresh series are not
distinguishable from most O. a. muelleri. In north-west Zambia, O. a. minuscula
White, 1946, was described as similar to ‘polyzona’, but smaller, the centre of the
breast deep rufous (like fuscata) and the belly very pale, almost whitish (White
1946). White (1963) later listed minuscula as a synonym of mulleri [sic]. In semi-
arid northern Botswana and the Hwange area of western Zimbabwe, O. a. pallida
are ‘paler above and below than O. a. bradfieldi Roberts [1929] (Roberts 1932).

At Lake Bangweulu, Mweru Marsh, Lake Kako and Abercorn (Mbala) in the
floodplain region of north-east Zambia, Benson (1955) described O. a. smithersi as
dark above, almost like fuscata. O. a. smithersi has broad white spectacles and a
white chin like muelleri, with rich rufous underparts and mostly black upperparts,
with broad black streaks and the grey streaks less extensive than in fuscata which it
most closely resembles in size. The bill of O. a. smithersi is ‘mainly sepia’ rather
than red in the dry season (July-August), perhaps non-breeders; in this region the
only breeding record of quailfinch is during the rains in February (Benson 1955).

Other plumage traits that differ between geographic populations of Ortygospiza
include the intensity of the underparts coloration, and the width of barring on the
breast and flanks. None of these traits varies distinctly between taxa (Table 1),
except for the darker and more boldly barred O. a. fuscocrissa in Ethiopia,
compared with quailfinch in adjacent regions. Size does not differ significantly in
the samples available, except that O. a. fuscocrissa has longer wings and black-
chinned O. a. gabonensis has shorter wings than the other measured quailfinch (Fry
2004; RBP unpubl.).

Bill colour

Bill colour was formerly reported to differ between black-chinned and white-
chinned populations of Ortygospiza (Traylor 1963, White 1963). Benson (1955)
proposed that this feature could be used to distinguish two species, red-billed O.
gabonensis and dark-billed O. atricollis. In fact, during the breeding season all adult
quailfinch photographed or with annotated specimens have red not dark bills,
regardless of taxon and geographic location.

Bill colour changes with season; breeding-season males have the upper
mandible bright red like the lower mandible (Immelmann ef al. 1965, 1977a,

Robert B. Payne & Michael D. Sorenson 10 Bull. B.O.C. 2007 127(1)

Traylor & Parelius 1967). Ten breeding males photographed or collected by RBP in
1995 in northern Nigeria (Jos, Bukuru), and a male in 1999 in The Gambia
(Marakissa) and another in Dalaba, Guinea (O. a. atricollis and O. a. ansorgei,
respectively) had bills ranging from partly red to uniformly red (Fig. 1). In Ethiopia,
specimens of O. a. fuscocrissa have the bill red in November (FMNH 83878) and
black in February (FMNH 83874); in birds taken in May, Heuglin (1863) described
the bill as blackish above (‘rostro nigricante’). In Kenya, van Someren collected two
O. a. muelleri at Lake Nakuru with a large white chin spot, white eye-ring, large
testes (one bird), and uniformly ‘coral red’ bill (FMNH 203787 in October, FMNH
257714 in December). In Tanzania the holotype of O. a. muelleri was a male with a
bright red bill (‘leuchtend rot’; Zedlitz 1911). In aviaries, male O. a. muelleri
observed as long as eight years had red bills in each breeding season (RBP); and
Ruschin (1972) observed red bills year-round in east African O. a. muelleri. In the
field, breeding-season birds in Natal, Zambia, Zimbabwe and Botswana also have
red bills (Clancey 1965; M. P. S. Irwin in litt. 2000), as they do in other regions of
Africa.

Juvenile quailfinch have dark bills. As the birds mature, the bills turn reddish,
first on the lower mandible (O. a. atricollis, Garoua and Ngaoundéré, Cameroon,
UMMZ 202407, 232473-75; O. a. fuscata, Angola, FMNH 84299, 84300; O. a.
gabonensis, Congo-Brazzaville, FMNH 213747; observations of non-breeding and
breeding O. a. muelleri in aviaries: RBP unpubl.).

Allopatry or sympatry of quailfinch taxa

Taxa of quailfinch are mainly allopatric. Near-sympatry between white-chinned
(‘Ortygospiza atricollis’) and black-chinned (‘O. gabonensis’) quailfinch has been
reported in four regions, yet none of these published reports involved known
breeding sympatry. Quailfinch in some regions are seasonal in their local occurrence
on floodplains and grasslands, and in the dry season they appear in areas where they
are not known to breed. As a result of their seasonal movements, different taxa
sometimes occur together.

Chapin (1954) reported two taxa in eastern Ituri, DR Congo, but he recognised
only one species, O. atricollis. Chapin’s report served as the recent basis for
concluding that quailfinch comprise two sympatric species (Traylor 1963, 1968,
Dickinson 2003). In fact, the birds were taken at different localities. On the
Albertine escarpment near Bogoro the birds (dorsostriata) lacked white on the face
and had the back more streaked black than birds at lower altitudes at Kasenyi
(ugandae), as at Kasindi and elsewhere west of the Rift and south of Lake Edward.
Birds at Bogoro were breeding in September; birds at Kasenyi and west of Lake
Albert and Lake Edward were not breeding in January and May (AMNH, BMNH,
FMNH). Bogoro specimens include one with ten feathers white at the base of the
chin (AMNH 264434); in plumage intermediate between dorsostriata and ugandae.

The taxon ugandae has been considered a synonym of dorsostriata (Sclater
1930a, Friedmann & Loveridge 1937). The holotype of ugandae (FMNH 257709)

Robert B. Payne & Michael D. Sorenson 1] Bull. B.O.C. 2007 127 (1)

TABLE 1
Plumage and bill characters of quailfinch taxa (male)'.

Subspecies Region Chin Throat Eye-ring Lores Back Back Flanks Belly Bill
white white white white colour streaked colour colour colour
ansorgei West +,(0) 0 0,(+) 0 dark no chestnut chestnut red
atricollis West-Central T aF 0,+ 0(+) brown slight chestnut chestnut red
ugandae upper Nile, E T + (0) + 0,4 brown slight tawny tawny -
dorsostriata upper Nile, W 06) 0) 0, 0 brown some tawny tawny -
gabonensis W equatorial 0 0 0 0 dark yes whitish whitish red
fuscata S-WC 0 0 0 0 blackish yes tawny chestnut red
smithersi N Zambia aE JHF + ++ blackish yes tawny chestnut -
Juscocrissa Ethiopia AMF IAF ack ++ brown some tawny tawny red
muelleri S to EC +++ i o eee tot | DROW slight buff buff red
pallida SC ++ dat amr tat pale slight buff pale buff red
brown-grey
digressa SE IAF Fal a Tar dark slight buff buff red
brown-grey

'In some taxa characters vary within a region; in Table 1 this variation is represented by two symbols, separated by
a comma; that in parentheses is the less common; “during breeding season.

from North Kavirondo has a small white chin spot and an incomplete, narrow white
eye-ring, which extends around the lores to the bill. It otherwise is nearly identical
to a bird with no white on the face (FMNH 118268) from Entebbe, identified as
dorsostriata; its plumage is intermediate but more like dorsostriata than O. a.
muelleri from southern Kenya (FMNH 257714). Van Someren (1922) noted that
ugandae has white around the eye and on the chin, whereas dorsostriata does not;
but male dorsostriata sometimes has a little white on the chin, and the back is more
distinctly streaked. Finally, on the north shore of Lake Victoria near Entebbe some
are intermediate in colour and pattern to the described taxa. Most specimens of
quailfinch in Uganda have no notation of large gonads on their labels; and there is
no evidence from either field observations or specimens that two taxa breed
assortatively in sympatry (Chapin 1954: 500; AMNH, FMNH, BMNH).

In Uganda both O. gabonensis and O. atricollis were reported at Semliki
Wildlife Reserve in a birding guide (Rossouw & Sacchi 1998), but not in the
Uganda bird atlas (Carswell et al. 2005), which remarked on the difficulty of
distinguishing these two forms. M. Wilson (Semliki contributor to Rossouw &
Sacchi 1998) has seen only black-chinned birds there; and when he and RBP
observed birds in August 2006, only black-chinned birds were seen on Semliki
Flats. Semliki is west of the eastern escarpment of Lake Albert; Murchison National
Park is east of the same escarpment. These observations do not support the
occurrence of two taxa at the same locality. In Uganda one series of reports appears
to refer to a single population with continuous plumage variation (van Someren

Robert B. Payne & Michael D. Sorenson 2 Bull. B.O.C. 2007 127(1)

1921a,b, 1922). In Uganda no seasonal breeding records are known for black-
chinned quailfinch, and only one record (June) for white-chinned quailfinch (Brown
& Britton 1980, Carswell et al. 2005).

Other published reports of sympatric quailfinch are from Zambia, but none
involved documented breeding or local sympatry in the breeding season. In north-
east Zambia, Benson (1955) noted two taxa of quailfinch, but not at the same
locality during the breeding season. O. a. smithersi was breeding in the south
Bangweulu region in February; both O. a. smithersi and O. a. fuscata were taken at
Abercorn (Mbala), the former on seasonal drying floodplains and the latter on
permanently wet grasslands or sponge dambos, and Benson suggested the birds
have different habitat preferences.

Second, in north-central Zambia near Ndola and the North Kafue basin, both
black-chinned and white-chinned birds have been reported, but the identifications of
white-chinned birds are in question. In this region, black-chinned quailfinch are
common, and all birds observed in the field and aviaries in this region were black-
chinned (Benson & Irwin 1967). At Itawa, Ndola, District Commissioner and
resident collector E. L. Button noted for fuscata, on the label of specimen FMNH
206576 taken on 3 September 1944, they have ‘been in just over a week, now
plentiful, found nest with incubated eggs in February and in March’. Penry (1986)
found only O. a. fuscata breeding at Chingola near Ndola.

Third, in north-west Zambia both white-chinned muelleri [‘polyzona’| and
black-chinned fuscata occur in November (Benson 1960), but quailfinch do not
breed there until the rains in January (White 1946, Benson et al. 1971). In the same
region Traylor (1963) reported black-chinned and white-chinned birds, but they
were in moult and not breeding at this time, November, and were taken in different
areas; fuscata on the damp floodplain of the perennial South Lueti River, muelleri
on the Liuwa Plain with only scattered surface water at the end of the dry season
(Traylor 1965; FMNH).

Other evidence of movements by quailfinch populations in certain areas is their
seasonal occurrence and absence. Quailfinch are regarded as local residents in some
areas, near permanent water at the edge of the Kafue Flats, Lochinvar National Park,
southern Zambia (Dowsett 1966), but at Mazabuka, within 10 km of the Kafue Flats
and 50 km of Lochinvar, they are seasonal visitors during the rains (Winterbottom
1959). They are absent near Choma, southern Zambia, in June—October (Aspinwall
1980). White-chinned quailfinch in Zambia are more widespread and liable to
‘considerable local movements depending on habitat conditions’ than black-chinned
quailfinch, though for the latter too ‘some minor seasonal movement remains
probable’ (Benson et al. 1971). In Malawi they undertake local movements and in
some areas are seen only in flocks (Benson 1953, Dowsett-Lemaire & Dowsett
2006). In parts of Zimbabwe they undergo ‘seasonal wandering’ (Irwin 1981), and
they are absent in some seasons in the highlands of south-east South Africa (Clancey
1996). In Kenya they are either resident or seasonal, as they appear in some areas

Robert B. Payne & Michael D. Sorenson 13 Bull. B.O.C. 2007 127(1)

during or after a wet season (both flocks and possible breeders) (Lewis & Pomeroy
1989).

In Zambia and east Africa, the plumages of male quailfinch taken in regions
between white-chinned and black-chinned populations are intermediate. (1) In
north-west Zambia (Barotseland) on the Liuwa Plain, birds taken by Traylor (1965;
FMNH) and first reported as minuscula are intermediate between the paler-backed,
less streaked muelleri of southern Zambia and northern Botswana, and the darker,
more streaked fuscata in north-west Zambia. Traylor’s birds have the white chin of
muelleri but the white eye-ring is narrower. (2) As in Benson (1955), smithersi of
north-east Zambia is a mosaic intermediate between the white-chinned ‘polyzona’
[=muelleri| to the south and the black-chinned fuscata to the north. (3) The
indistinctly streaked back and the incomplete and narrow white eye-ring of ugandae
in Uganda are intermediate between traits of dorsostriata in the west and muelleri
in the east. A few recognised as dorsostriata in Uganda at Mpumu and Kigambo
have some white on the chin (Cowles 1957; BMNH). Sclater (1930a) considered
ugandae a synonym of dorsostriata, whereas Cowles (1957) and Rand et al. (1959)
suggested that dorsostriata is a synonym of gabonensis. (4) In the north-west Congo
and Gabon, birds identified as gabonensis include one specimen with white on the
chin (Cowles 1957). More field work may reveal additional information about
movements and local variation of quailfinch populations.

In summary, of the records of two taxa of quailfinch in sympatry, in neither Ituri
or in Uganda were they in local breeding sympatry. In north-west Zambia they were
not in local sympatry and were not seen together in the breeding season, and in
north-east Zambia birds were taken in different habitats and not during the breeding
season (mainly January—March: Benson et al. 1971, Fry 2004), and some may have
been seasonal non-breeding visitors. Quailfinch are mainly allopatric, and only in
parts of Zambia were they said to ‘occur on the same ground’ (Britton 1980), but not
in the breeding season. There is no direct information that the white-chinned and
black-chinned quailfinch of east, central and southern Africa co-occur without
interbreeding. Quailfinch are seasonal in occurrence in some areas, absent until the
rains, then appear and breed; they are locally migratory, and sometimes occur
outside their breeding area. Furthermore, those in both north-east and north-west
Zambia (smithersi and ‘minuscula’) are intermediate between the nearest other
populations to the north and south.

mtDNA phylogeny

We obtained sequences of the mitochondrial ND2 gene for 12 specimens
representing nearly all recognised taxa across the geographic range of the species
(Table 2). We used the following criteria to select specimens for genetic
information: 1) the most recent specimens that were available for molecular
sampling; mtDNA deteriorates with age and the more recent specimens can be
amplified and sequenced with greater accuracy (Payne & Sorenson 2003, Sefc et al.

Robert B. Payne & Michael D. Sorenson 14 Bull. B.O.C. 2007 127(1)

2003, 2006); 2) in North American museums; and 3) voucher specimens were
compared with other series to validate the identification.

Laboratory methods were identical to previous studies (Sorenson et al. 2004,
Sorenson & Payne 2005) except that estrildid-specific internal primers were used to
permit the amplification and sequencing of smaller DNA fragments from older
specimens. Primer pairs for tissue samples were L5216rv and H5766rv (Sefc et al.
2003), and L5758rv (5’°-GGNGGNTGAATRGGNYTNAAYCARAC-3’) and
H6313rv (5’-ACTCTTRTTTAAGGCTTTGAAGGC-3’). Additional internal
primers included L5476.E (TTYKCYAGYATRAYYAAYGCATG), H5481
(TGNGTRATRTCYCAYTGDCCNGT), L6007.E (TCHCTNGCAGGNY TNCCNCC),
and H6022.E (GTHAGTTCTTGGATGATNAGTCATTTTGG); primer names refer
to the strand and position of the 3’ base in the Gallus gallus mtDNA sequence
(Desjardins & Morais 1990). ND2 sequences for the three Amandava species and
two Amadina species were used as the outgroup. Phylogenetic analysis based on
parsimony and maximum likelihood (ML) produced identical results. The latter
analysis used a general time-reversible model of nucleotide substitution with an
estimated proportion of invariant sites; model chosen based on AIC value as
calculated in MODELTEST (Posada & Crandall 1998) using parameters estimated
from the data. Genetic distances reported below are ML estimates based on the same
model and parameter values.

These data produced a single well-supported monophyletic tree (Fig. 2) with
three distinct and genetically divergent clades. (1) A clade including west African

O. a. ansorgei, The Gambia

99/70
O. a. aincollis, Cameroon

O. a. ugandae, W Kenya
©. a. gabonensis , Gabon
O. a. fuscata, NW Zambia

100/99 | 4.4% 100/100

O. a. dorsosinata, S Uganda

O. a. dorsostrata, E Uganda

99/100 | 2.5% f ue
O. a. fuscocnssa, Ethiopia

O. a. mueller’, S Zambia
O. a. muelleri, captive
100/97 O. a. digressa, 5 Mozambique

O. a. smithersi, N Zambia

0.01 substitutions/site

Figure 2. Phylogeny of quailfinch mitochondrial DNA lineages based on complete sequences of the ND2
gene (outgroup taxa not shown). Branch lengths are proportional to maximum likelihood (ML) estimates
of number of substitutions per nucleotide. The mean ML genetic distance across the two basal nodes is
shown. Parsimony and ML bootstrap values are shown for the primary groups on the tree.

Robert B. Payne & Michael D. Sorenson 15 Bull. B.O.C. 2007 127(1)

TABLE 2
Sources and voucher specimens of genetic samples.

Taxon Locality Year Tissue? Voucher specimen
ansorgei Marakissa, The Gambia 1996 l UMMZ 234175
atricollis Ngaoundere, Cameroon 1992 l UMMZ 232472
ugandae Mumias, North Kavirondo, western Kenya 1917 R FMNH 257709
muelleri captive (parents known) 1993 l UMMZ 233156
muelleri Lochinvar National Park, southern Zambia 1972 2 UMMZ 219735
digressa Beira, Mozambique 1965 2 UMMZ 211483
fuscocrissa Gojam, Ethiopia 1927 2 FMNH 83874
dorsostriata Entebbe, eastern Uganda 1915 2 UMMZ 94816
dorsostriata Masaka, southern Uganda 1916 2 FMNH 91697
fuscata Kawambwa, northern Zambia 1953 2 UMMZ 222394
gabonensis Ogoouć-Maritime, Gabon 1951 2 FMNH 210584
smithersi Abercorn, northern Zambia 1954 3 AMNH 648202

è Tissue used for genetic analysis. 1: muscle tissue; 2: single feather from specimen; 3: toe pad tissue from
specimen.

ansorgei, atricollis and ugandae is basal to the other two. (2) Black-chinned taxa in
central Africa, gabonensis, fuscata and dorsostriata, form a second clade, and (3)
the white-chinned taxa occurring from eastern to southern Africa (fuscocrissa,
muelleri, smithersi and digressa) form a third clade. Relatively large genetic
distances between clades (2.5-4.4%) combined with limited intra-clade
differentiation suggest long-term historical isolation of quailfinch populations in
different regions, dating perhaps a million years or more (e.g. Fleischer et al. 1998,
Arbogast et al. 2002, 2006).

The results on the basis of genetic monophyly are consistent with the possibility
of one, two or three species of quailfinch. However, if two species were recognised,
these would be west African O. atricollis (Vieillot, 1817) (clade 1 in the gene tree)
and central and east African O. fuscocrissa Heuglin, 1863 (the oldest-named taxon
in clades 2 and 3 in Fig. 2), and this clade would include black-chinned gabonensis.
These two clades do not correspond to any previously suggested arrangement for
quailfinches. The main conclusion from the genetic data is that we reject the
recognition of two species, one with a white chin and one with a black chin. The
three clades are consistent with the three Rassengruppen of a single species of
quailfinch as described by Wolters (1975) and recently recognised as three species
(Fry 2004). The mostly ‘white-chinned’ quailfinch of west Africa are basal to a
clade comprising black-chinned quailfinch of central Africa and spectacled
quailfinch of eastern and southern Africa. In conclusion, the genetic data are
consistent both with one species (O. atricollis) and with three species (O. atricollis,
O. fuscocrissa, O. gabonensis), and do not support the hypothesis of two species (O.
atricollis and O. gabonensis).

Robert B. Payne & Michael D. Sorenson 16 Bull. B.O.C. 2007 127(1)

Song

Vocalisations of quailfinch include a short contact-call, a harsh klek that rises
quickly to 3-4 kHz and persists for c.0.07—1.0 sec, and a song (here, ‘loudsong’)
characterised by a harsh pattern of irregularly alternating notes given in phrases that
repeat with some variation, as klik klak kloik kluk klek (Immelmann et al. 1965,
Goodwin 1982, Nuttall 1993, Payne & Payne 1994). The loudsongs continue for up
to several seconds. The loud k/ek contact-calls and klik klak kloik loudsongs develop
from modulated calls given by young after they fledge and are heard in flocks with
juvenile quailfinch (Payne & Payne 1994). Female loudsongs are short and discrete
(Nuttall 1993; RBP unpubl.), whereas male loudsong phrases repeat over time with
minor variations of the sequence of notes in the klik klak kloik motifs (Fig. 3;
compare with Nuttall 1993, Fig. 5, where labeled as ‘contact phrase’ rather than
‘song’). These same elements of loudsong are also given as contact-calls and rattling
flight-calls. Loudsongs are usually given on the ground. In RBP’s aviaries, the birds
began loudsong well before bright lights-on or sunrise. Shorter versions of loudsong
also are given on the ground in the field and in the flight aviaries where the birds
bred successfully. In Fig. 3, according to the field recordists’ notes
(http://www.bl.uk/collections/sound-archive/cat.html, viewed 9 November 2006),
songs a, b, d, e, f and j were of perched birds, as probably was song k (the recordist
‘saw it well’). Songs c, g, h and i lack data on context. In both form and context,
most vocalisations in Fig. 3 are loudsongs, either complete or incomplete (contra
Nuttall 1993, 2005, who did not recognise these as ‘song’); the other vocalisations
consist of the same kinds of shorter calls. Calls and songs have the same acoustic
elements in several other estrildids as well (e.g.,. Immelmann 1969, Nicolai 1964,
Gittinger 1970, Zann 1975, 1976, Payne & Payne 1994).

Loudsongs of quailfinch are similar throughout their distribution. The klik klak
kloik... songs of ‘white-chinned’ birds in The Gambia, Nigeria and Cameroon
(Barlow et al. 2002; British Library Natural Sound Archive [NSA]; RBP),
Murchison National Park, Uganda (RBP), southern Zambia (Stjernstedt 1993; NSA)
and South Africa (Nuttall 1993; NSA), and in ‘black-chinned’ birds in northern
Zambia (Stjernstedt 1994; NSA) are all similar. The loudsongs have hoarse notes
given in irregular sequence, sometimes the first higher and the last lower; one long,
the second mid length, and the third short (Fig. 3). Penry’s (1986) take it away and
drink descriptions of flight-calls of black-chinned birds in northern Zambia apply
equally to white-chinned O. a. muelleri elsewhere in Zambia and to white-chinned
O. a. atricollis in Cameroon and Nigeria (Payne & Payne 1994; Fig. 3). Take it away
is also a motif in the loudsong klik klak kloik. In Ethiopia, the songs of white-
chinned O. a. fuscocrissa were described in similar terms, ‘ . . . die Stimme ist ein
ziemlich unmelodisches Pipen, das oft schnarrend klingt und weit weniger fein, als
bei den vorhergehenden’ (Heuglin 1863). In north-west Zambia, when a black-
chinned male quailfinch ‘O. gabonensis’ was collected, ‘the call did not differ in any
way from that of atricollis’ (S. Keith in Benson & Irwin 1965). Quailfinch in the

Robert B. Payne & Michael D. Sorenson 17 Bull. B.O.C. 2007 127(1)

a +
’ 4 a |
« a m we lon
tte TE etal $e
= i <a l = of et ws t E e e
5 { a

ar ern Same fo, a peepee ma a p eee
p le
hi n
ia. wt
L ariak
n
t “he
kn
=
_*
ETT °*
Ąą %
ch
TY "E
=x
FO

1 j = m i i n ¥ W W
W rgy Wwe F iha wa t hi p
ii yet METT,
Lie E uT Be thy | y
2| ]

0 1.8 sec

Figure 3. Loudsongs of quailfinch Ortygospiza atricollis. (a) O. a. ansorgei, captive (UMMZ 232576);
(b) O. a. atricollis <> ansorgei, Sifoe, The Gambia (Barlow et al. 2002; UMMZ 234175); (c-d) O. a.
atricollis, c, Zaria, Nigeria (NSA 3635); d, Rayfield, Nigeria, 3 November 1995 (RBP 49A); (e-g) O. a.
muelleri, e, male b/b, captive (RBP 1996 tape 70b, UMMZ 236155); f, female r/r, captive (RBP 1991
tape 20b, UMMZ 234138); g, South Africa, Kruger National Park (Gillard 1987); (h-j) O. a. fuscata,
Itawa, Ndola region, northern Zambia (h, NSA 26681; i, NSA 26692; j, NSA 81087; k, NSA 25769).

Robert B. Payne & Michael D. Sorenson 18 Bull. B.O.C. 2007 127(1)

Ndola region of northern Zambia (including Chingola and Itawa Swamp) were
identified as black-chinned fuscata. Recordings of these birds were examined (NSA
34109, 34110, 25769, 26681, 26692, 81085, 81087). A bird recorded at Itawa
Swamp on 4 May 1991 was ‘re-identified’ as (white-chinned) ‘O. atricollis’ on an
edited copy of NSA 25769; and this edited copy is suspect, both for acoustic
parameters and for the re-identification. Chappuis (2000) reported that calls of
black-chinned birds at Itawa, near Ndola, Zambia, were more shrill than white-
chinned birds, but no difference is evident in the unedited NSA recording or in
published cassettes (Stjernstedt 1993, 1994). The modulated ‘shrill’ sounds in
certain recordings (not all calls are shown in Fig. 3) appear to be calls of juveniles
(Payne & Payne 1994).

The pitch of calls may vary with excitement level. Penry’s (1986) idea of
species-distinctive pitch and modulation in calls of black-chinned and white-
chinned quailfinch is not supported by his audiospectrograms, nor do differences
appear in larger series of recordings, where both black-chinned and white-chinned
birds give loudsongs either over or under 4 kHz (Fig. 3). Song recordings of
quailfinch vary in the presence and loudness of an undertone of the loud trace at 4
kHz. The occurrence of sounds above and below the fundamental frequency (the
loudest trace on the audiospectrogram) appear to vary with loudness of the call, the
excitement of the bird, and the recording conditions (overtones often occur in ‘over-
recorded’ samples with settings of recorder sensitivity producing artefacts, and
undertones are not prominent in birds recorded at close range in captivity).
Additionally, the undertones appear more often in notes of 4 kHz or higher than in
lower notes. The overtone and undertone amplitudes at different frequencies are
easily distorted, and the relative amplitudes may vary with recording conditions
(Wickstrom 1982). Acoustic harmonics vary with a male’s distance to his mate not
only in quailfinch (Nuttall 1993) but also in another estrildid, Poephila acuticauda
(Zann 1975). This modulation of song overtones may be effected by active neural
control of resonance filtering (Beckers et al. 2003). In consequence, we do not
emphasise the acoustic overtones and undertones; we merely point to their
occurrence and prominence in the quailfinch.

A third vocalisation is a soft ‘burbling’ song of irregularly repeated downslurred
notes of short (<0.05 s) duration. This soft “burbling’ or ‘scissors-grinding’ song is
used in sexual behaviour at close range and at the nest. Soft song is delivered at an
amplitude c.10—15 dB lower than loudsongs (Payne & Payne 1994). Soft songs are
similar in west African and southern African quailfinch (O. a. ansorgei and O. a.
atricollis, and O. a. muelleri) (Nuttall 1993, Payne & Payne 1994); no recordings of
soft songs are available for black-chinned quailfinch.

Nestling mouth pattern and colour

Nestling mouth markings have been used to diagnose closely related estrildid
finches (Nicolai 1987). Nestlings and fledglings have distinctive mouth patterns and
colours they display to the adult when begging for food. These may offer visual

Robert B. Payne & Michael D. Sorenson 19 Bull. B.O.C. 2007 127(1)

signals that aid in behavioural recognition and parental care. In some estrildids,
young with atypical mouths receive less food and survive less well than young with
the species-typical mouth patterns and colours (Immelmann et al. 1977b, Payne et
al. 2001). For this reason, nestlings with different mouth patterns and colours might
not have the visual signals necessary for parental care. If nestling mouth patterns
and colours differed between birds, lower fitness of ‘hybrid’ offspring might present
a barrier to successful interbreeding (Payne 2005).

In nestling and fledgling quailfinch the mouth has three greenish-blue balls each
side of the gape, a black gape between them, and a yellowish palate with six black
spots, a pattern of colours and spots that differs from all other estrildids (Payne
2005). Recently fledged young in a family group of O. a. atricollis photographed in
Cameroon at Ngaoundere in 1992 (UMMZ) have the same mouth colours and
patterns as nestlings throughout the range of white-chinned quailfinch in Nigeria (O.
a. atricollis), Kenya and South Africa (O. a. muelleri) (Serle 1938, van Someren
1956, Schifter 1964, Kunkel 1966, Nuttall 1992, Payne & Payne 1994, Payne 2005).
Mouths and palates of young O. a. fuscata of the black-chinned gabonensis complex
are apparently the same as in the other known quailfinch (Chapin 1954, Payne
2005). There is no evidence of different nestling begging signals and mouth colours
between quailfinch taxa that would affect whether a brood is reared successfully.

Discussion

To assess systematic status of quailfinch taxa, we combine information on
morphology, distribution, behaviour and molecular genetics as analysed in a
phylogenetic context. The uniformity across quailfinch taxa in song and bill colour
of adults, and in the mouth colours and patterns of nestlings, indicates a single
species, Ortygospiza atricollis. In addition, no geographically neighbouring taxa of
quailfinch are known to breed assortatively in sympatry, whereas in many cases they
intergrade morphologically: atricollis with ansorgei and ugandae; ugandae with
dorsostriata; smithersi with fuscata and muelleri, and fuscata with ‘minuscula’ and
muelleri. The migratory behaviour of some populations may lead them to co-occur
in the non-breeding season, but reported cases of sympatry involve birds that breed
in different localities; for example, in Ituri, breeding populations of ugandae and
dorsostriata are 50 km apart (Chapin 1954). Plumage variation in Ortygospiza
(Table 1) indicates only one or two character differences between neighbouring taxa
in most instances. The gradation of plumage between taxa, the lack of known local
breeding sympatry, and the consistency of songs across taxa are consistent with
previous conclusions of a single species (Friedmann & Loveridge 1937, Chapin
1954, Benson 1955, Traylor & Parelius 1967). These reports, like those of museum
workers (Sclater 1930a, Wolters 1975, 1985), recognised quailfinch as a single
species because the variation in plumage pattern and colour between the black- and
white-chinned forms is no greater than variation between subspecies.

With genetic data, one approach to the recognition of species is the genetic
distance between clades: the percentage of nucleotide sites that differ between

Robert B. Payne & Michael D. Sorenson 20 Bull. B.O.C. 2007 127(1)

sequences, sometimes based on a single gene and on an arbitrary ratio of the
variation within and between nominal species (Hebert et al. 2004). However, there
is no agreement in genetic distance across sister species of birds. For example, the
arbitrary threshold noted above is poor in predicting species status in other
estrildids, in which family we find numerous examples of both small genetic
distances between sister species (e.g., <1.0%) and deep genetic distances within a
species (up to 6%; Sorenson et al. unpubl.). Genetic distances vary greatly among
different lineages and the level of difference between species in one lineage does not
predict the difference in other lineages (DeSalle et al. 2005). In other animals, the
error rate of recognising or not recognising known species on the basis of genetic
distance is as great as 30% (Meyer & Paulay 2005, Hickerson et al. 2006, Meier et
al. 2006). |

A phylogenetic analysis of genetic data in quailfinch suggests the recognition of
either one or three clades as species. In general, though many clades are species, not
all are so (i.e., some clades distinguish families and genera, e.g. Sorenson & Payne
2005, whereas at the other end of the scale, some mitochondrial lineages simply
trace maternal kinship, e.g. Avise 2004, Payne et a/. 2002). Furthermore, not all
genes differ between related species (Avise 2004). For example, in Vidua finches,
several species share the same mitochondrial gene sequences, due to incomplete
lineage sorting between species that derived from a polymorphic ancestor, and to
hybridisation (Sorenson et al. 2003). In Poephila finches, different nuclear genes
provide discordant estimates of phylogenetic relationships between species.
Nevertheless, mitochondrial gene trees are concordant with a majority of nuclear
gene trees from the same birds (Jennings & Edwards 2005). In addition, although
ND2 is slightly more variable than ND3 and Cyt-d, their nucleotide sequences give
nearly identical phylogenetic results in other songbirds (Zink & Weckstein 2003).
These observations support the use of mtDNA at the species level in the finches,
with one variable sequence (ND2) being representative of other such sequences
(Jennings & Edwards 2005, Edwards et al. 2005).

Within the framework of an integrative species concept, the more important
question is whether genetic differences are congruent with morphological and/or
behavioural differences between taxa. Additional information is necessary before
we can determine the biological and evolutionary significance of a clade (Avise
2004). For example, does a clade correspond to birds with the same morphology, the
same songs, and a geographic distribution that overlaps that of a closely related
population—three criteria that are necessary to consider in determining at which
level a clade corresponds to a species. Because collectively the quailfinch are
monophyletic, they form a pattern of ancestry and descent, a generally necessary
condition but not a sufficient condition for recognising a species. Certain clades are
recognisable within the collective clade, yet the lack of a mutually exclusive and
morphologically diagnosable set of morphological characters (Table 1) suggests it
is inappropriate to recognise any one included clade as a species. The idea of two
quailfinch species is unsupported by the genetic data, because the lineages in

Robert B. Payne & Michael D. Sorenson 21 Bull. B.O.C. 2007 127(1)

‘white-chinned’ taxa are paraphyletic, as are the taxa in ‘atricollis? as recognised by
White (1963). Based on his interpretation of our genetic phylogeny, Fry (2004)
divided quailfinch into three species diagnosable by the presence or absence of a
few key plumage characters; conspicuous white spectacles (‘O. fuscocrissa’), a lack
of white on the chin and face (‘O. gabonensis’), and presence of a white chin but no
prominent white on the face (‘O. atricollis’). However, these characters vary both
within and between the three lineages recovered in a phylogenetic analysis of
mtDNA nucleotide sequences (Table 1, Fig. 2), and there is morphological evidence
of gene flow in regions where the different plumage taxa almost meet. Although the
genetic data suggest that quailfinch populations were isolated in three different
regions for a significant part of their history, the contact in current distributions and
observations of birds intermediate in plumage do not support a conclusion that the
three clades correspond to species-level lineages that are on independent
evolutionary trajectories. Of course the tree topology also suggests distinct genetic
lineages or independent evolutionary pathways, but we cannot know the future or
whether the lineages will combine again before they become extinct; we can only
interpret the past from data that link the birds at the present time (de Queiroz &
Donoghue 1988, O’Hara 1994, de Queiroz 1998).

When results from single genes yield unresolved phylogenetic trees, it is
desirable to determine the phylogenies of multiple genes. In the present case, the
gene tree based on ND2 gene sequences is well supported, and adding more mtDNA
data is unlikely to change our inferences. It might be of interest to recover genetic
sequence data within a large sample; that is, to test the robustness of the two most
basal branching points in the phylogenetic estimate in Fig. 2. And data from nuclear
loci would be of interest, but many specimens in our study range in age up to 90
years. Obtaining nuclear sequence data from these older specimens would require a
great deal of additional effort and expense; and, given the relatively limited
sampling in the field, particularly from regions of contact between subspecies, it is
not clear to us that these data would add much to the present analysis. More
extensive and intensive sampling in zones of overlap would be helpful, as with O.
a. ugandae to test its association with O. a. atricollis, O. a. dorsostriata and O. a.
muelleri: recently collected specimens are simply unavailable. Additional
geographic sampling and analysis of nuclear loci also would provide a better
understanding of the historical structure and current extent of genetic exchange
between populations, and a large sample would be necessary to distinguish between
variation within and between taxa. Presumably, geographic isolation of quailfinch
populations in the past explains the three divergent genetic clades. Though
subsequent expansion and shifts in distribution have probably brought them back
into contact in certain areas, their contemporary distributions are largely non-
overlapping.

The morphological variation within and among quailfinch taxa, the absence of
documented breeding sympatry, the seasonal movements of birds between breeding
and non-breeding areas, and the similarity in songs all suggest gene flow between

Robert B. Payne & Michael D. Sorenson 22 Bull. B.O.C. 2007 127(1)

adjacent populations and the lack of divergence in signaling systems that would
occur in speciation. Together with the molecular phylogeny these observations lead
us to conclude that quailfinch are best recognised as a single, geographically
variable species, O. atricollis.

Acknowledgements

For access to specimens we thank the American Museum of Natural History, New York (AMNH),
Natural History Museum, Tring (BMNH), Durban Museum, Field Museum of Natural History, Chicago
(FMNH), National Museum of Natural History, Washington DC (USNM), Muséum National d’ Histoire
Naturelle, Paris (MNHN), and National Museums of Zimbabwe, Bulawayo. For digital photographs and
notes on specimens we thank K. Cook (BMNH), J.-F. Voisin (MNHN), H. van Grouw at the Nationaal
Natuurhistorisch Museum, Leiden (RMNH), and I. Heynen at the Staatliches Museum für Naturkunde
Stuttgart (SMNS). For comments on quailfinch and species we thank C. R. Barlow, C. W. Benson, R. J.
Dowsett, F. Dowsett-Lemaire, C. H. Fry, M. P. S. Irwin, L. L. Knowles and M. Wilson. FMNH provided
genetic samples of O. a. gabonensis, O. a. dorsostriata and O. a. ugandae, AMNH a sample of O. a.
smithersi, and the other samples are from University of Michigan Museum of Zoology, Ann Arbor
(UMMZ). Four reviewers commented on the manuscript. Copies of calls and songs were provided by the
British Library National Sound Archive, London (NSA) through Richard Ranft and Joanne Nicholson,
by C. Chappuis, F. Dowsett-Lemaire and C. R. Barlow, and RBP recorded birds in the field and in
aviaries.

References:

Alström, P. & Ranft, R. 2003. The use of sounds in bird systematics, and the importance of bird sound
archives. Pp. 113—135 in Collar, N. J., Fisher, C. & Feare, C. J. (eds.) Why museums matter: avian
archives in an age of extinction. Bull. Brit. Orn. CI. 123A.

Arbogast, B. S., Edwards, S. V., Wakeley, J., Beerli, P. & Slowinski, J. B. 2002. Estimating divergence
times from molecular data on phylogenetic and population genetic timescales. Ann. Rev. Ecol. &
Syst. 33: 707-740.

Arbogast, B. S., Drovetski, S. V., Curry, R. L., Boag, P. T., Seutin, G, Grant, P. R., Grant, B. R. &
Anderson, D. J. 2006. The origin and diversification of Galapagos mockingbirds. Evolution 60:
370-382.

Aspinwall, D. R. 1980. Field observations. Zambian Orn. Soc. Newsletter 10(2): 35-37.

Avise, J. C. 2004. Molecular markers, natural history, and evolution. Second edn. Sinauer, Sunderland,
MA.

Bannerman, D. A. 1949. The birds of tropical west Africa, vol. 7. Crown Agents, London.

Barlow, C. & Wacher, T. 1997. A field guide to birds of The Gambia and Senegal. Pica Press,
Robertsbridge.

Barlow, C. R., Hammick, J. & Seller, P. 2002. Bird song of The Gambia and Senegal. CD. Mandarin
Productions, Robertsbridge.

Bates, G. L. 1930. Handbook of the birds of west Africa. John Bale, Sons & Danielsson, London.

Beckers, G. J. L., Suthers, R. A. & ten Cate, C. 2003. Pure-tone birdsong by resonance filtering of har-
monic overtones. Proc. Natl. Acad. Sci. USA 100: 7372-7376.

Benson, C. W. 1953. A check list of the birds of Nyasaland. Nyasaland Soc. & Publications Bureau,
Blantyre & Lusaka.

Benson, C. W. 1955. New forms of pipit, longclaw, robin-chat, grass-warbler, sunbird, quail-finch and
canary from central Africa. Bull. Brit. Orn. Cl. 75: 101-109.

Benson, C. W. 1960. Recent records from north-western Northern Rhodesia, part 2. Bull. Brit. Orn. CI.
80: 114-119.

Benson, C. W., Brooke, R. K., Dowsett, R. J. & Irwin, M. P. S. 1971. The birds of Zambia. Collins,
London.

Benson, C. W. & Irwin, M. P. S. 1965. Some birds from the North-Western Province, Zambia. Arnoldia
(Rhod.) 29: 1-11.

Robert B. Payne & Michael D. Sorenson 23 Bull. B.O.C. 2007 127(1)

Benson, C. W. & Irwin, M. P. S. 1967. A contribution to the ornithology of Zambia. National Museums
of Zambia & Oxford Univ. Press.

Borrow, N. & Demey, R. 2001. A guide to the birds of western Africa. Princeton Univ. Press.

Britton, P. L. (ed.) 1980. Birds of east Africa. E. Afr. Nat. Hist. Soc., Nairobi.

Brown, L. H. & Britton, P. L. 1980. The breeding seasons of east African birds. E. Afr. Nat. Hist. Soc.,
Nairobi.

Carswell, M., Pomeroy, D., Reynolds, J. & Tushabe, H. 2005. The bird atlas of Uganda. British
Ornithologists’ Club & British Ornithologists’ Union, Oxford.

Cassin, J. 1860. Catalogue of birds collected on the Rivers Camma and Ogobai, western Africa, by Mr.
P. B. Duchaillu, in 1858, with notes and descriptions of new species, continued. Proc. Acad. Nat. Sci.
Phil. 1859: 133-144.

Chapin, J. P. 1954. The birds of the Belgian Congo, part 4. Bull. Amer. Mus. Nat. Hist. 75B: 1-846.

Chappuis, C. 2000. African bird sounds, vol. 2. Société Ornithologique de France, Paris & British Library
National Sound Archive, London.

Clancey, P. A. 1958. Miscellaneous taxonomic notes on African birds XI. Durban Mus. Novit. 5:
117-142.

Clancey, P. A. 1965. The South African races of the quail finch. Arnoldia (Rhod.) 2: 1—6.

Clancey, P. A. 1977. Miscellaneous taxonomic notes on African birds XLVII. Durban Mus. Novit. 11:
181-212.

Clancey, P. A. 1996. The birds of southern Mozambique. Revised edn. African Bird Club Publishing,
Westville, Kwazulu-Natal.

Clement, P. 1993. Finches and sparrows: an identification guide. Princeton Univ. Press.

Cowles, G. S. 1957. A note on Ortygospiza gabonensis Lynes. Bull. Brit. Orn. Cl. 77: 60—61.

Cracraft, J. 1983. Species concepts and speciation analysis. Current Orn. 1: 159-187.

Dayrat, B. 2005. Towards integrative taxonomy. Biol. J. Linn. Soc. 85: 407—415.

de Queiroz, K. 1998. The general lineage concept of species, species criteria, and the processes of speci-
ation: a conceptual unification and terminological recommendations, Pp. 57-75 in Howard, D. &
Berlocher, S. H. (eds.) Endless forms: species and speciation. Oxford Univ. Press.

de Queiroz, K. 2005. Ernst Mayr and the modern concept of species. Proc. Natl. Acad. Sci. USA 102:
6600—6607.

de Queiroz, K. & Donoghue, M. J. 1988. Phylogenetic systematics and the species problem. Cladistics
4: 317-338.

DeSalle, R., Egan, M. G. & Siddall, M. 2005. The unholy trinity: taxonomy, species delimitation and
DNA barcoding. Phil. Trans. Roy. Soc. Lond. B 360: 1905-1916.

Desjardins, P. & Morais, R. 1990. Sequence and gene organization of the chicken mitochondrial genome.
A novel gene order in higher vertebrates. J. Mol. Biol. 212: 599-634.

Dickinson, E. C. (ed.) 2003. The Howard & Moore complete checklist of the birds of the world. Third
edn. Christopher Helm, London.

Dowsett, R. J. 1966. A preliminary list of the birds of the Kafue Flats. Puku 4: 101—124.

Dowsett-Lemaire, F. & Dowsett, R. J. 2006. The birds of Malawi. Tauraco Press & Aves, Liege.

Edwards, S. V., Kingan, S. B., Calkins, J. D., Balakrishnan, C. N., Jennings, W. B., Swanson, W. J. &
Sorenson, M. D. 2005. Speciation in birds: genes, geography, and sexual selection. Proc. Natl. Acad.
Sci. 102 suppl. 1: 6550—6557.

Fleischer, R. C., McIntosh, C. E. & Tarr, C. L. 1998. Evolution on a volcanic conveyor belt: using
phylogeographic reconstructions and K-Ar-based ages of the Hawaiian Islands to estimate molecular
evolutionary rates. Mol. Ecol. 7: 533-545.

Friedmann, H. & Loveridge, A. 1937. Notes on the ornithology of tropical east Africa. Bull. Mus. Comp.
Zool. 81: 1-413.

Fry, C. H. 2004. Genus Ortygospiza Sundevall. Pp. 383-390 in Fry, C. H. & Keith, S. (eds.) The birds of
Africa, vol. 7. Christopher Helm, London.

Gatter, W. 1997. Birds of Liberia. Pica Press, Robertsbridge.

Gillard, L. 1987. Southern African bird calls, part 3. Gillard Bird Cassettes, Johannesburg.

Robert B. Payne & Michael D. Sorenson 24 Bull. B.O.C. 2007 127(1)

Goodwin, D. 1982. Estrildid finches of the world. Brit. Mus. (Nat. Hist.), London.

Grant, C. H. B. & Mackworth-Praed, C. W. 1956. On the status of the quail finch (Ortygospiza polyzona
(Temminck)). Ostrich 27: 41.

Groth, J. G. 1998. Molecular phylogenetics of finches and sparrows: consequences of character state
removal in cytochrome b sequences. Mol. Phyl. & Evol. 10: 377-390.

Güttinger, H. R. 1970. Zur Evolution von Verhaltensweisen und Lautausserungen bei Prachtfinken
(Estrildidae). Zeitschrift fur Tierpsychologie 27: 1011-1075.

Hebert, P. D. N., Stoeckle, M. Y., Zemlak, T. S. & Francis, C. M. 2004. Identification of birds through
DNA barcodes. PLoS Biol. 2: 1657-1663.

Heuglin, T. von. 1863. Beiträge zur Ornithologie Nord-Ost Afrika’s. J. Orn. 11: 3-29.

Hickerson, M. J., Meyer, C. P. & Moritz, C. 2006. DNA barcoding will often fail to discover new animal
species over broad parameter space. Syst. Biol. 55: 729-739.

Immelmann, K. 1969. Song development in the zebra finch and other estrildid finches. Pp. 61—81 in
Hinde, R. A. (ed.) Bird vocalizations. Cambridge Univ. Press.

Immelmann, K., Steinbacher, J. & Wolters, H. E. 1965. Prachtfinken, vol. 1. Second, revised edn. Verlag
Hans Limberg, Aachen.

Immelmann, K., Nicolai, J., Steinbacher, J. & Wolters, H. E. 1977a. Prachtfinken, vol. 2. Second, revised
edn. Verlag Hans Limberg, Aachen.

Immelmann, K., Piltz, A. & Sossinka, R. 1977b. Experimentelle Untersuchungen zur Bedeutung der
Rachenzeichnungen junger Zebrafinken. Zeitschrift für Tierpsychologie 45: 210-218.

Irwin, M. P. S. 1981. The birds of Zimbabwe. Quest, Harare.

Jennings, W. B. & Edwards, S. V. 2005. Speciational history of Australian grass finches (Poephila)
inferred from thirty gene trees. Evolution 59: 2033-2047.

Keay, R. W. J. 1959. Vegetation map of Africa south of the Tropic of Cancer. Oxford Univ. Press.

Knowles, L. L. & Carstens, B. In press. Delimiting species without monophyletic gene trees. Syst. Biol.

Kunkel, P. 1966. Bemerkung zu einigen Verhaltensweisen des Rebhuhnastrilds, Ortygospiza atricollis
atricollis (Vieillot). Zeitschrift fur Tierpsychologie 23: 136—140.

Lewis, A. & Pomeroy, D. 1989. A bird atlas of Kenya. A. A. Balkema, Rotterdam.

Linnaeus, D. 1758. Systema naturae, vol. 1. Tenth edn. L. Salvii, Holmiae.

Mayr, E. 1963. Animal species and evolution. Oxford Univ. Press.

Mayr, E. 2000. The biological species concept. Pp. 17—29 in Wheeler, Q. D. & Meier, R. (eds.) Species
concepts and phylogenetic theory. Columbia Univ. Press, New York.

Mayr, E., Paynter, R. A. & Traylor, M. A. 1968. Family Estrildidae. Pp. 306-389 in Paynter, R. A. (ed.)
Check-list of birds of the world, vol. 14. Mus. Comp. Zool., Cambridge, MA.

Meier, R., Shiyang, K., Vaidya, G. & Ng, P. K. L. 2006. DNA barcoding and taxonomy in Diptera: a tale
of high intraspecific variability and low identification success. Syst. Biol. 55:715-728.

Meyer, C. P. & Paulay, G 2005. DNA barcoding: error rates based on comprehensive sampling. PLoS
Biol. 3: 2229-2238.

Moreau, R. E. 1966. The bird faunas of Africa and its islands. Academic Press, London.

Moritz, C. & Cicero, C. 2004. DNA barcoding: promise and pitfalls. PLoS Biol. 2: 1529-1531.

Nicolai, J. 1964. Der Brutparasitismus der Viduinae als ethologisches Problem. Pragungsphanomene als
Faktoren der Rassen- und Artbildung. Zeitschrift für Tierpsychologie 21: 129-204.

Nicolai, J. 1987. Die Rachenzeichnungen von Großer und Kleiner Ptinktchenamarant (Lagonosticta
nitidula und L. rufopicta). Trochilus 8: 116-120.

Nikolaus, G. 1987. Distribution atlas of Sudan’s birds with notes on habitat and status. Bonn. Zool.
Monogr. 25.

Nuttall, R. J. 1992. Breeding biology and behaviour of the quail finch Ortygospiza atricollis. Ostrich 63:
110-117.

Nuttall, R. J. 1993. Vocal behaviour of the quail finch Ortygospiza atricollis. Ostrich 64: 97—104.

Nuttall, R. J. 2005. African quailfinch Ortygospiza atricollis. Pp. 1040—1041 in Hockey, P. A. R., Dean,
W. R. J. & Ryan, P. G. (eds.) Roberts birds of South Africa. Seventh edn. John Voelcker Bird Book
Fund, Cape Town.

Robert B. Payne & Michael D. Sorenson 25 Bull. B.O.C. 2007 127(1)

Ogilvie-Grant, W. R. 1910. [Ortygospiza ansorgei sp. nov.] Bull. Brit. Orn. CI. 25: 84.

O’Hara, R. J. 1994. Evolutionary history and the species problem. Amer. Zoologist 34: 12-22.

Payne, R. B. 2005. Nestling mouth markings and colors of Old World finches Estrildidae: mimicry and
coevolution of nesting finches and their Vidua brood parasites. Univ. Mich. Mus. Zool. Misc. Publ.
194.

Payne, R. B., Hustler, K., Stjernstedt, R., Sefc, K. M. & Sorenson, M. D. 2002. Behavioural and genetic
evidence of a recent population switch to a novel host species in brood parasitic indigobirds Vidua
chalybeata. Ibis 145: 373-383.

Payne, R. B. & Payne, L. L. 1994. Song mimicry and species associations of west African indigobirds
Vidua with quail-finch Ortygospiza atricollis, goldbreast Amandava subflava and brown twinspot
Clytospiza monteiri. Ibis 136: 291—304.

Payne, R. B., Payne, L. L., Woods, J. L. & Sorenson, M. D. 2000. Imprinting and the origin of parasite-
host species associations in brood parasitic indigobirds Vidua chalybeata. Anim. Behav. 59: 69-81.

Payne, R. B. & Sorenson, M. D. 2003. Museum collections as sources of genetic data. Pp. 97—104 in
Rheinwald, G. (ed.) Bird collections in Europe: the challenge of mutual cooperation. Bonn. Zool.
Beitr. 51.

Payne, R. B., Woods, J. L. & Payne, L. L. 2001. Parental care in estrildid finches: experimental tests of
a model of Vidua brood parasitism. Anim. Behav. 62: 473-483.

Penry, E. H. 1986. Notes on the biology of the redbilled quailfinch (Ortygospiza gabonensis) in Zambia.
Ostrich 57: 193-202.

Posada, D. & Crandall, K. A. 1998. Modeltest: testing the model of DNA substitution. Bioinformatics 14:
817-818.

Rand, A. L., Friedmann, H. & Traylor, M. A. 1959. Birds from Gabon and Moyen Congo. Fieldiana Zool.
41: 221-411.

Roberts, A. 1929. New forms of African birds. Ann. Transvaal Mus. 13: 71-81.

Roberts, A. 1930. Some recent important records of South African birds. Ostrich 1: 61-68.

Roberts, A. 1932. Preliminary descriptions of sixty-six new forms of South African birds from the
Vernay-Lang Expedition, 1932. Ann. Transvaal Mus. 15: 21-33.

Rossouw, J. & Sacchi, M. 1998. Where to watch birds in Uganda. Uganda Tourist Board, Kampala.

Ruschin, H. W. 1972. Etwas tiber Wachtelastrilde, ihre Haltung und Zucht. Gefiederte Welt 98: 141.

Schifter, H. 1964. Weitere Zuchterfolge mit Wachtelastrilden. Gefiederte Welt 90: 83-85.

Sclater, W. L. 1930a. Systema avium A:thiopicarum. British Ornithologists’ Union, London.

Sclater, W. L. 1930b. [Ortygospiza atricollis fuscata, subsp. nov.]. Bull. Brit. Orn. Cl. 52: 142-143.

Searcy, W. A. & Yasukawa, K. 1996. Song and female choice. Pp. 454—473 in Kroodsma, D. E. & Miller,
E. H. (eds.) Ecology and evolution of acoustic communication in birds. Cornell Univ. Press, Ithaca,
NY.

Sefc, K. M., Payne. R. B. & Sorenson, M. D. 2003. Microsatellite amplification from museum feather
samples: the effects of fragment size and template concentration on genotyping errors. Auk 120:
982-989.

Sefc, K. M., Payne, R. B. & Sorenson, M. D. 2006. Single base errors in PCR products from avian muse-
um specimens and their effect on estimates of historical genetic diversity. Conserv. Genetics
doi.org/10.1007/s10592-006-9240-8.

Serle, W. 1938. Observations on the breeding habits of Nigerian estrildine weaver-birds. Oologists’ Rec.
18: 40-45.

Serle, W. & Morel, G J. 1977. A field guide to the birds of west Africa. Collins, London.

Sharpe, R. B. 1890. Catalogue of the birds in the British Museum, vol. 13. Trustees Brit. Mus. (Nat.
Hist.), London.

Shelley, G. E. 1905. The birds of Africa, vol. 4, part 1. R. H. Porter, London.

Sites, J. W. & Marshall, J. C. 2004. Operational criteria for delimiting species. Ann. Rev. Ecol., Evol. &
Syst. 35: 199-227.

Sorenson, M. D., Balakrishnan, C. N. & Payne, R. B. 2004. Clade-limited colonization in brood parasitic
finches (Vidua spp.). Syst. Biol. 53: 140-153.

Robert B. Payne & Michael D. Sorenson 26 Bull. B.O.C. 2007 127(1)

Sorenson, M. D. & Payne, R. B. 2005. A molecular genetic analysis of cuckoo phylogeny. Pp. 68—94 in
Payne, R. B. The cuckoos. Oxford Univ. Press.

Sorenson, M. D., Sefc, K. M. & Payne, R. B. 2003. Speciation by host switch in brood parasitic indigo-
birds. Nature 424: 928-931.

Stjernstedt, R. 1993. Birdsong of Zambia, part III. Tape cassette. R. Stjernstedt, Lusaka.

Stjernstedt, R. 1994. Rare birds of Zambia. Tape cassette. R. Stjernstedt, Lusaka.

Temminck, C. J. 1838. Nouveau recueil des planches coloriées d’oiseaux, vol. 3. F. G. Levrault, Paris.

Traylor, M. A. 1963. Revision of the quail finch Ortygospiza atricollis. Bull. Brit. Orn. Cl. 83: 141-146.

Traylor, M. A. 1965. A collection of birds from Barotseland and Bechuanaland. /bis 107: 137—172,
357-384.

Traylor, M. A. 1968. Estrildidae (African). Pp. 306-389 in Paynter, R. A. (ed.) Check-list of birds of the
world, vol. 14. Mus. Comp. Zool., Cambridge, MA.

Traylor, M. A. & Parelius, D. 1967. A collection of birds from the Ivory Coast. Fieldiana Zool. 51:
91-117.

van Someren, V. G. L. 1921a. [Ortygospiza atricollis dorsostriata, subsp. nov.] Bull. Brit. Orn. Cl. 41:
115.

van Someren, V. G. L. 1921b. [Ortygospiza atricollis ugandae, subsp. nov.] Bull. Brit. Orn. CI. 41: 121.

van Someren, V. G. L. 1922. Notes on the birds of east Africa. Novit. Zool. 29: 1—247.

van Someren, V. G. L. 1956. Days with birds. Fieldiana Zool. 38: 1-520.

Vieillot, L. J.-P. 1817. Nouveau dictionnaire d’histoire naturelle. New edn. Deterville, Paris.

Wheeler, Q. D. & Platnick, N. I. 2000. The phylogenetic species concept (sensu Wheeler and Platnick).
Pp. 55—69 in Wheeler, Q. D. & Meier, R. (eds.) Species concepts and phylogenetic theory. Columbia
Univ. Press, New York.

White, C. M. N. 1946. The ornithology of the Kaonde-Lunda Province, Northern Rhodesia. Part 4. [bis
88: 206-224.

White, C. M. N. 1963. A revised check list of African flycatchers, tits, tree creepers, sunbirds, white-eyes,
honey eaters, buntings, finches, wavers and waxbills. Govt. Printer, Lusaka.

Wickstrom, D. C. 1982. Factors to consider in recording avian sounds. Pp. 1—52 in Kroodsma, D. E. &
Miller, E. H. (eds.) Acoustic communication in birds, vol. 1. Academic Press, New York.

Will, K. W., Mishler, B. D. & Wheeler, Q. D. 2005. The perils of DNA barcoding and the need for inte-
grative taxonomy. Syst. Biol. 54: 844-851.

Winterbottom, J. M. 1959. Notes on the status of some birds in Northern Rhodesia. Ostrich 30: 1-12.

Wolters, H. E. 1975. Die Vogelarten der Erde. Paul Parey, Berlin.

Wolters, H. E. 1985. Species limits in some Afrotropical Estrildidae (Aves: Passeriformes). Proc. Intern.
Symp. Afr. Vertebrates, Bonn 1985: 425—434.

Zann, R. 1975. Inter and intraspecific variation in the calls of the three species of grassfinches of the sub-
genus Poephila (Gould) (Estrildidae). Zeitschrift für Tierpsychologie 39: 85-125.

Zann, R. 1976. Variation in the songs of three species of estrildine grassfinches. Emu 76: 97-108.

Zedlitz, O. 1911. Nachtrége zu ‘Meine Ornithologische Ausbeute in Nordost-Afrika’. J. Orn. 59:
591-613.

Zink, R. M. & Weckstein, J. D. 2003. Recent evolutionary history of the fox sparrows (genus:
Passerella). Auk 120: 522-527.

Addresses: Prof. Robert B. Payne, Museum of Zoology and Department of Ecology and Evolutionary
Biology, University of Michigan, Ann Arbor, Michigan 48109-1079, USA. Prof. Michael D.
Sorenson, Department of Biology, Boston University, 5 Cummington Street, Boston, Massachusetts
02215, USA.

© British Ornithologists’ Club 2007